Mr.Anime API

1. Create a token on your account page (API tokens card). It is shown once — copy it.
2. Send it as a bearer token on every request.

curl -H "Authorization: Bearer ma_live_YOUR_TOKEN" \
  "https://mr-anime.vercel.app/api/v1/anime/search?q=naruto&perPage=5"

// JavaScript
const res = await fetch("https://mr-anime.vercel.app/api/v1/anime/trending?perPage=10", {
  headers: { Authorization: `Bearer ${process.env.MRANIME_API_TOKEN}` },
});
const { data, pageInfo } = await res.json();

Every request needs a token in the Authorization header as Bearer ma_live_… (an X-API-Key header works too). Tokens start with ma_live_, are stored hashed server-side, and only the SHA-256 hash is kept — losing the token means revoking it and creating a new one. Revocation takes effect immediately.

Rate limit: 60 requests per minute per token. Over-limit responses are 429 with a Retry-After header; every success carries X-RateLimit-Remaining. Responses are CDN-cached for 10 minutes to 1 hour, so repeat reads are cheap.

CORS is enabled (Access-Control-Allow-Origin: *), so the API can be called directly from browser apps.

Mr.Anime is also an OAuth 2.0 provider — register an app on your account page (OAuth applications card) to get a client_id and a one-time client_secret. Two grant types are supported:

• authorization_code — “Sign in with Mr.Anime”: users approve your app, you receive a single-use 5-minute code, then exchange it for a 30-day access token.
• client_credentials — server-to-server API access with no user (read scope only).

https://mr-anime.vercel.app/oauth/authorize
  ?client_id=CLIENT_ID
  &redirect_uri=YOUR_CALLBACK
  &response_type=code
  &scope=read%20profile
  &state=RANDOM_UNGUESSABLE_STRING

curl -X POST https://mr-anime.vercel.app/oauth/token \
  -H "Content-Type: application/x-www-form-urlencoded" \
  -d grant_type=authorization_code \
  -d code=CODE \
  -d redirect_uri=YOUR_CALLBACK \
  -d client_id=CLIENT_ID \
  -d client_secret=CLIENT_SECRET

# → { "access_token": "ma_at_…", "token_type": "Bearer",
#     "expires_in": 2592000, "scope": "read profile" }

curl -X POST https://mr-anime.vercel.app/oauth/token \
  -H "Content-Type: application/x-www-form-urlencoded" \
  -d grant_type=client_credentials \
  -d client_id=CLIENT_ID \
  -d client_secret=CLIENT_SECRET

# → { "access_token": "ma_at_…", "expires_in": 2592000, "scope": "read" }

curl -H "Authorization: Bearer ma_at_YOUR_TOKEN" \
  "https://mr-anime.vercel.app/api/v1/me"   # or any anime endpoint

# → { "data": { "id": "…", "username": "…", "avatarUrl": null,
#               "role": "user", "isVerified": false, "badges": [] } }

List endpoints return data (an array of anime) plus pageInfo; the detail endpoint returns a single anime under data. The anime shape is identical everywhere:

GET /api/v1/anime/16498

{
  "data": {
    "id": 16498,
    "type": "ANIME",
    "title": { "romaji": "Shingeki no Kyojin", "english": "Attack on Titan", "native": "進撃の巨人" },
    "format": "TV",
    "status": "FINISHED",
    "episodes": 25,
    "duration": 24,
    "season": "SPRING",
    "seasonYear": 2013,
    "startDate": { "year": 2013, "month": 4, "day": 7 },
    "endDate": { "year": 2013, "month": 9, "day": 29 },
    "score": 84,
    "meanScore": 84,
    "popularity": 16800000,
    "favourites": 320000,
    "genres": ["Action", "Drama", "Fantasy"],
    "description": "Centuries ago, mankind was slaughtered to near extinction…",
    "cover": { "extraLarge": "https://s4.anilist.co/…", "large": "https://s4.anilist.co/…", "color": "#e4784c" },
    "banner": "https://s4.anilist.co/…",
    "studios": ["Wit Studio"],
    "nextAiringEpisode": null,
    "isAdult": false
  }
}

GET /api/v1/anime/trending?perPage=2

{
  "data": [ { "id": 16498, "type": "ANIME", "title": { … }, … } ],
  "pageInfo": { "currentPage": 1, "hasNextPage": true, "total": 500 }
}

/anime/{id}/episodes mirrors the site’s episode grid — titles and thumbnails come from TMDB first, then AniList’s streaming list:

GET /api/v1/anime/16498/episodes

{
  "data": [
    { "number": 1, "title": "To You, in 2000 Years", "thumbnail": "https://image.tmdb.org/t/p/w780/…" },
    { "number": 2, "title": "That Day", "thumbnail": "https://image.tmdb.org/t/p/w780/…" }
  ],
  "meta": { "episodes": 25, "season": 1 }
}

/anime/{id}/episodes/{episode} mirrors the watch page: full metadata, resolved episode info and one ready-to-embed url per provider (in switcher order), with per-provider dub availability probed the same way the site does:

GET /api/v1/anime/16498/episodes/1?lang=dub

{
  "data": {
    "anime": { … the same object as GET /anime/{id} … },
    "episode": { "number": 1, "title": "To You, in 2000 Years", "thumbnail": "https://image.tmdb.org/t/p/w780/…" },
    "season": 1,
    "requestedLang": "dub",
    "effectiveLang": "sub",
    "dubAvailable": false,
    "dubMissing": true,
    "providers": [
      { "id": "provider-1", "name": "Megaplay", "url": "https://…/1/sub", "dubAvailable": false },
      { "id": "provider-2", "name": "Anixo",    "url": "https://…/1",     "dubAvailable": true }
    ]
  }
}

providers[].url can be loaded directly in an iframe (srcdoc-free plain embed). If you request lang=dub but no provider has dub for that episode, effectiveLang becomes sub and dubMissing is true — exactly what the site shows.

Errors always use this envelope:

{
  "error": { "code": "invalid_token", "message": "The provided API token is not valid." }
}